A model-adaptive evolutionary algorithm for optimization

Many applications in engineering and science rely on the optimization of computationally expensive functions. A successful approach in such scenarios is to couple an evolutionary algorithm with a mathematical model which replaces the expensive function. However, models introduce several difficulties, such as their inherent inaccuracy, and the difficulty of matching a model to a particular problem. To address these issues, this paper proposes a model-based evolutionary algorithm with two main implementations: (a) it combats model inaccuracy with a tailored trust-region approach to manage the model during the search, and to ensure convergence to an optimum of the true expensive function, and (b) during the search it continuously selects an optimal model type out of a set of candidate models, resulting in a model-adaptive optimization search. Extensive performance analysis shows the efficacy of the proposed algorithm.     

Protease nexin 1 in the murine kidney: glomerular localization and up-regulation in glomerulopathies

Protease nexin 1 (PN-1), a potent serpin-class antiprotease, is thought to be synthesized in the murine kidney. However, neither the cellular localization of PN-1 synthesis nor its role has yet been defined. To address these questions, we determined by in situ hybridizations RNase protection assay and immunoblotting, the sites of PN-1 mRNA accumulation in normal mouse kidneys and the modulation of PN-1 expression in several pathological conditions. In normal kidneys, PN-1 mRNA was detected primarily in glomeruli, most likely in mesangial cells. The glomerular expression of PN-1 was substantially enhanced not only in lupus-like glomerulonephritis (induced by IgG3 monoclonal rheumatoid factors or occurring spontaneously in lupus-prone mice), but also in mild glomerular lesions associated with intracapillary thrombi induced by IgG3 anti-trinitrophenyl monoclonal antibodies. In contrast, no modulation of PN-1 mRNA levels was observed during the course of lipopolysaccharide-induced acute tubular necrosis. A constitutive PN-1 gene expression and its up-regulation during glomerular injury suggest a possible role for PN-1 in glomerular biology. In view of its high inhibitory activity towards thrombin, mesangial PN-1 may be involved in the control of glomerular coagulation following initial glomerular injuries.     

Contributions of Ea(z) and Eb(z) MHC genes to lupus susceptibility in New Zealand mice

Membrane vesicles derived from external taste epithelia of channel catfish (Ictalurus punctatus) were incorporated into lipid bilayers on the tips of patch pipettes. Consistent with previous experiments (Teeter, J. H., J. G. Brand, and T. Kumazawa. 1990. Biophys. J. 58:253-259), micromolar (0.5-200 microM) concentrations of L-arginine (L-Arg), a potent taste stimulus for catfish, activated a nonselective cation conductance in some bilayers, which was antagonized by D-Arg. Two classes of L-Arg-gated receptor/channels were observed in reconstituted taste epithelial membranes: one with a unitary conductance of 40-60 pS, and the other with a conductance of 75-100 pS. A separate class of nonselective cation channels, with a conductance of 50-65 pS, was activated by high concentrations of L-proline (L-Pro) (0.1-3 mM), which is the range necessary to elicit neural responses in catfish taste fibers. The L-Pro-activated channels were not affected by either L- or D-Arg, but were blocked by millimolar concentrations of D-Pro. Conversely, neither L- nor D-Pro altered the activity of either class of L-Arg-activated channels, which were blocked by micromolar concentrations of D-Arg. These results are consistent with biochemical, neurophysiological, and behavioral studies indicating that taste responses of channel catfish to L-Arg are mediated by high-affinity receptors that are part of or closely coupled to nonselective cation channels directly gated by low concentrations of L-Arg, while responses to L-Pro are mediated by distinct, low-affinity receptors also associated with nonselective cation channels.     

Induction of apoptosis in human pancreatic carcinoma cells by a synthetic bleomycin-like ligand

The increased polylactosamine glycosylation of LAMP-2 in MDCK cells cultured for 1 day relative to cells cultured for 3 days has been correlated with its slower rate of Golgi transit (Nabi and Rodriguez-Boulan, 1993, Mol. Biol. Cell., 4, 627-635). To determine if the differential polylactosamine glycosylation of LAMP-2 is a consequence of glycosyltransferase expression levels, the activities of beta1-6GlcNAc-TV, beta1-3GlcNAc-T(i), beta1-2GlcNAc-TI, beta1, 4Gal-T, alpha2-6sialyl-T, and alpha2-3sialyl-T were assayed and no significant differences in the activities of these enzymes in 1 and 3 day cell extracts were detected. During MDCK epithelial polarization, the Golgi apparatus undergoes morphological changes and apiconuclear Golgi networks were more evident in 3 day cells. Treatment with nocodazole disrupted Golgi networks and generated numerous Golgi clusters in both 1 day and 3 day cells. In the presence of nocodazole the differential migration of LAMP-2 in 1 and 3 day MDCK cells was maintained and could be eliminated by treatment with endo-beta-galactosidase, indicating that gross Golgi morphology did not influence the extent of LAMP-2 polylactosamine glycosylation. Nocodazole treatment did, however, result in the faster migration of LAMP-2 which was not due to modification of core N-glycans as the precursor form of the glycoprotein migrated with an identical molecular size. Following incubation at 20 degrees C, which prevents the exit of proteins from the trans-Golgi network, the molecular size of LAMP-2 increased to a similar extent in both 1 and 3 day MDCK cells. Extending the time of incubation at 20 degrees C did not influence the size of LAMP-2, demonstrating that its glycosylation is modified not by its retention within the Golgi but rather by its equivalent slower Golgi passage at the lower temperature in both 1 and 3 day cells. An identical effect was observed in nocodazole treated cells, demonstrating that Golgi residence time determines the extent of LAMP-2 polylactosamine glycosylation, even in isolated Golgi clusters.     

Dielectric characters of 0.7Pb(Mg1/3Nb2/3)O3–0.3PbTiO3 ceramics fabricated at ultra-low temperature by the spark-plasma-sintering method

Solid solution (1 − x)Pb(Mg_1/3Nb_2/3)O₃–xPbTiO₃ shows high dielectric constant near room temperature and is an ideal capacitor material. The composition 0.7Pb(Mg_1/3Nb_2/3)O₃–0.3PbTiO₃, which is located near the morphotropic phase boundary, were densified by the spark-plasma-sintering method at an ultra-low temperature (700 °C). Dielectric constant measurement shows that the thus prepared sample shows higher dielectric constant at room temperature and good temperature stability in a wide temperature range. The behavior is much different from that of samples sintered by conventional method and could be ascribed to size effect.     

Tubular up-regulation of clusterin mRNA in murine lupus-like nephritis

Clusterin, a widely distributed glycoprotein, is detected in most tissues and in numerous physiological fluids. In the kidney, this protein is constitutively expressed in tubular epithelial cells, and its expression is enhanced following tubular injuries. In addition, clusterin has been detected in glomerular immune deposits of glomerulonephritis. The present study was designed to define the sites of clusterin mRNA accumulation in murine lupus-like nephritis in comparison with murine tubulopathies. In lupus-like nephritis, a significant increase of clusterin mRNA abundance was demonstrated. This up-regulation was localized exclusively in tubular epithelial cells exhibiting tubulointerstitial alterations, whereas no clusterin mRNA was detectable in diseased glomeruli, excluding an active synthesis of clusterin by glomerular cells. A similar tubular increase of clusterin mRNA abundance was observed in myeloma-like cast nephropathy induced by IgG3 monoclonal cryoglobulins and even in the absence of any detectable histological alterations in a model of septic shock induced by the injection of bacterial lipopolysaccharides. Our results suggest that tubular epithelial cells are the only sites of clusterin mRNA accumulation during the course of lupus-like nephritis and that the tubular up-regulation of clusterin gene expression may reflect the cellular response to various types of tubular injuries.     

Isogeometric topology optimization of anisotropic metamaterials for controlling high-frequency electromagnetic wave

This study presents a level set–based topology optimization with isogeometric analysis (IGA) for controlling high-frequency electromagnetic wave propagation in a domain with periodic microstructures (unit cells). The high-frequency homogenization method is applied to characterize the macroscopic high-frequency waves in periodic heterogeneous media whose wavelength is comparative to or smaller than the representative length of a unit cell. B-spline basis functions are employed for the IGA discretization procedure to improve the performance of electromagnetic wave analysis in a unit cell and topology optimization. Also, to keep the same order of continuity on the periodic boundaries as on other element edges in the domain, we propose the extended domain approach, while incorporating Floquet periodic boundary condition (FPBC). Two types of optimization problems are taken as examples to demonstrate the effectiveness of the proposed method in comparison with the standard finite element analysis (FEA). The optimization results provide optimized topologies of unit cells qualified as anisotropic metamaterials with hyperbolic and bidirectional dispersion properties at the macroscale.     

Decision-making support system for human resource allocation in product development projects

Recent dismal economic conditions and a ruthlessly competitive environment have forced many companies to restructure, or reorganize their priorities. For such companies, the concentration of various resources upon their particular corporate strong points has become a central strategy. Consequently, there has been a rapid increase in the importance of (1) selecting profitable projects from a wealth of possible alternatives and (2) optimizing the allocation of current resources among the selected projects. This paper proposes an optimization system for project selection that not only yields the most beneficial project set, but also the optimum allocation of human resources for the selected projects. The optimization system consists of two algorithms, namely (1) a project selection algorithm for choosing the set of projects that maximizes the total estimated profit, and (2) a human resource allocation algorithm for optimally placing human resources among the selected projects, having considered the satisfaction level provided by each employee's skills, personal motivation and career goals.     

A structural optimization method based on the level set method using a new geometry‐based re‐initialization scheme

Structural optimization methods based on the level set method are a new type of structural optimization method where the outlines of target structures can be implicitly represented using the level set function, and updated by solving the so‐called Hamilton–Jacobi equation based on a Eulerian coordinate system. These new methods can allow topological alterations, such as the number of holes, during the optimization process whereas the boundaries of the target structure are clearly defined. However, the re‐initialization scheme used when updating the level set function is a critical problem when seeking to obtain appropriately updated outlines of target structures. In this paper, we propose a new structural optimization method based on the level set method using a new geometry‐based re‐initialization scheme where both the numerical analysis used when solving the equilibrium equations and the updating process of the level set function are performed using the Finite Element Method. The stiffness maximization, eigenfrequency maximization, and eigenfrequency matching problems are considered as optimization problems. Several design examples are presented to confirm the usefulness of the proposed method. Copyright © 2010 John Wiley & Sons, Ltd.